As AI becomes more interwoven into our day-to-day lives, we will want to understand how it has been trained to perceive the world so we know what to expect. We may even want to have multiple personas available to us, as I might expect a different tone and voice from my life coach than I would from my personal assistant.

Personas can be defined up-front in the AI's initial training, or at the surface level in the form of subsequent prompts.

Foundational prompt

Each LLM on the market has a foundational prompt that guides the way it thinks about the world and interacts with users. Most of these are private, though Anthropic has shared theirs publicly. It's surprisingly short!

Because of this, the choice of your underlying model will have downstream impacts on the AI. Understanding how each LLM thinks is the first step towards defining the persona for your AI overall.

The foundational prompt for Claude vs. ChaptGPT could be intuited based on how each answers a simple question

Subsequent training

When you think about AI personalities, you're probably thinking about voice and tone, contextual awareness, and so on. These are traits that can be trained later on. When creating AIs for your company, think about the different scenarios where it would be playing, and what characteristics are helpful in each. A human-centered approach helps here. If I needed technical support, I want my agent to be helpful and not overly personal. That changes if the person is my professor, or my friend. Take a similar approach to developing AI agents. Some apps allow users to define these themselves, and examining the consumer side of this, like at, can help us think through what this looks like in a B2B context.

Comparing and Zendesk helps us find parallels in how these different platforms are approaching this shared problem

Details and variations

  • Offer default personalities out of the box and let users customize them
  • An AI's persona is different from the user's personal voice. The first case relates to how the AI interacts directly with users, the second relates to how the AI understand's the users' intent to speak or generate on their behalf



The right person for the job

AI personas help to make AI more personal and adaptable, which builds user trust and helps them get things done with less friction. Just as we use personality clues to assess the humans around us, expect users to do the same with their digital counterparts.

Potential risks

Brand fragmentation

There's a tradeoff to having too many characters or personas in a business setting. When done right, this can lead to more personal interactions. On the other hand, it can feel kitschy and lead to overall distrust of the use of AI. Treat all of your AI personas with care, not just the ones handling tough situations. Users will combine them in your head as a reflection of your overall brand.

Use when:
You need your AI to connect emotionally with the user or stand out in the market.


Individual instructions can be given to define the general persona of ChatGPT
While not explicitly mapped to personal characteristics, ChatGPT lets users pick the voice ChatGPT should use in voiceovers allows you to define personal characteristics to each persona
Users can interact with each persona individually or view them together in an aggregated sandbox highlights how the different personas of their AIs can work together, each on a different area of need
No items found.